AR_final file_2018-19

resonant detectors of GW, and also the correspond- ing probabilities of GW induced transitions that the phonon modes of such resonant detectors un- dergo. We present the complete perturbative calcu- lation involving both time-independent and time- dependent perturbations in the Hamiltonian. This work has been done in collaboration with Bhat- tacharyya, and Sunandan Gangopadhyay. Sanjay Kumar Sahay Group-wise classification approach to improve An- droid malicious apps detection accuracy In the fast-growing smart devices, Android is the most popular OS, and due to its attractive features, mobility, ease of use, these devices hold sensitive information such as personal data, browsing his- tory, shopping history, financial details, etc. There- fore, any security gap in these devices means that the information stored or accessing the smart de- vices are at high risk of being breached by the mal- ware. These malware are continuously growing, and are also used for military espionage, disrupt- ing the industry, power grids, etc. To detect these malware, traditional signature matching techniques are widely used. However, such strategies are not capable to detect the advanced Android malicious apps because malware developer uses several obfus- cation techniques. Hence, researchers are continu- ously addressing the security issues in the Android based smart devices. Therefore, in this work us- ing Drebin benchmark malware dataset, we exper- imentally demonstrate how to improve the detec- tion accuracy by analyzing the apps after group- ing the collected data based on the permissions and achieved 97.15% overall average accuracy. Our results outperform the accuracy obtained without grouping data. The analysis also shows that among the groups, Microphone group detection accuracy is least while Calendar group apps are detected with the highest accuracy, and for the best performance, one shall take 80-100 features. This work has been done in collaboration with Ashu Sharma. An investigation of a deep learning based malware detection system In the investigation, we experiment with differ- ent combination of deep learning architectures in- cluding Auto-Encoders, and Deep Neural Networks with varying layers over Malicia malware dataset, on which earlier studies have obtained an accuracy of (98%) with an acceptable False Positive Rates (1.07%). But these results were done using exten- sive man-made custom domain features and invest- ing corresponding feature engineering and design efforts. In this proposed approach, besides improv- ing the previous best results (99.21% accuracy and a False Positive Rate of 0.19%) indicates that deep learning based systems could deliver an effective defense against malware. Since it is good in au- tomatically extracting higher conceptual features from the data, deep learning based systems could provide an effective, general and scalable mecha- nism for detection of existing and unknown mal- ware. This investigation has been done in collabo- ration with Mohit Sewak, and Hemant Rathore. Sandeep Sahijpal A Monte Carlo based simulation of the galactic chemical evolution of the Milky Way galaxy The formation and chemical evolution of the Milky Way galaxy is numerically simulated by develop- ing a Monte Carlo approach to predict the elemen- tal abundance gradients and other galactic features using the revised solar abundance. The galaxy is accreted gradually by using either a two-infall or a three-infall accretion scenario. The galaxy is chemically enriched by the nucleosynthetic contri- butions from an evolving ensemble of generations of stars. We analyse the role of star formation ef- ficiency. The influence of the radial gas inflow as well as radial gas mixing on the evolution of galaxy is also studied. The SN Ia delay time distribution (DTD) is incorporated by synthesizing SN Ia pop- ulations using random numbers based on a distri- bution function. The elemental abundance evolu- tionary trends corroborate fractional contributions of ∼ 0.1 from prompt ( < 100 Myr) SN Ia popu- lation. The models predict steep gradients in the inner regions and less steep gradients in the outer regions, which agrees with the observations. The gradients indicate an average radial gas mixing ve- locity of ≤ 1 km/s − 1 . The models with radial gas inflows reproduce the observed inversion in the ele- mental abundance gradients around 2 billion years. The three-infall accretion scenario performs better than the two-infall accretion model in terms of ex- plaining the elemental abundance distributions of the galactic halo, thick and thin discs. The ac- curacy of all the models has been monitored as a c umulative error of < 0 . 15 M ⊙ in the mass bal- ( 212 )